fix: prevent saturation from RMS normalization clipping peaks

RMS normalize to target then scale back if peaks exceed 1.0,
preserving dynamics instead of hard-clipping transients.
Eval sample target updated to -23 dBFS to match training data.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

nodes/selva_lora_trainer.py

@@ -134,9 +134,12 @@ def _eval_sample(generator, feature_utils_orig, dataset, seq_cfg, device, dtype,
         elif audio.dim() == 3 and audio.shape[1] != 1:
             audio = audio.mean(dim=1, keepdim=True)
 
-        target_rms = 10 ** (-20.0 / 20.0)   # -20 dBFS
+        target_rms = 10 ** (-23.0 / 20.0)   # -23 dBFS matches training data
         rms = audio.pow(2).mean().sqrt().clamp(min=1e-8)
-        audio = (audio * (target_rms / rms)).clamp(-1, 1)
+        audio = audio * (target_rms / rms)
+        peak = audio.abs().max().clamp(min=1e-8)
+        if peak > 1.0:
+            audio = audio / peak
         return audio.squeeze(0), seq_cfg.sampling_rate   # [1, L]
 
     except Exception as e:

nodes/selva_sampler.py

@@ -174,7 +174,12 @@ class SelvaSampler:
         if normalize:
             target_rms = 10 ** (target_lufs / 20.0)
             rms = audio.pow(2).mean().sqrt().clamp(min=1e-8)
-            audio = (audio * (target_rms / rms)).clamp(-1, 1)
+            audio = audio * (target_rms / rms)
+            # If RMS normalization pushes peaks into clipping, scale back to
+            # preserve dynamics rather than hard-clipping (no saturation)
+            peak = audio.abs().max().clamp(min=1e-8)
+            if peak > 1.0:
+                audio = audio / peak
         print(f"[SelVA] audio: shape={tuple(audio.shape)} sr={sample_rate}", flush=True)
 
         return ({"waveform": audio.cpu(), "sample_rate": sample_rate},)